Matcha-TTS: A fast TTS architecture with conditional flow matching

TL;DR

Matcha-TTS is a fast, high-quality TTS system using an ODE-based decoder trained with optimal-transport conditional flow matching, achieving efficiency and quality without external alignments.

Contribution

Introduces Matcha-TTS, a novel encoder-decoder architecture utilizing OT-CFM for rapid, high-quality TTS synthesis with minimal memory footprint.

Findings

Outperforms baseline models in speed and quality.

Achieves highest mean opinion score in listening tests.

Has the smallest memory footprint among compared models.

Abstract

We introduce Matcha-TTS, a new encoder-decoder architecture for speedy TTS acoustic modelling, trained using optimal-transport conditional flow matching (OT-CFM). This yields an ODE-based decoder capable of high output quality in fewer synthesis steps than models trained using score matching. Careful design choices additionally ensure each synthesis step is fast to run. The method is probabilistic, non-autoregressive, and learns to speak from scratch without external alignments. Compared to strong pre-trained baseline models, the Matcha-TTS system has the smallest memory footprint, rivals the speed of the fastest models on long utterances, and attains the highest mean opinion score in a listening test. Please see https://shivammehta25.github.io/Matcha-TTS/ for audio examples, code, and pre-trained models.